The long-term storage of carbon in the deep ocean by sinking particles is a key piece of the global carbon cycle, playing a critical role in setting atmospheric CO2 levels, driving ocean de-oxygenation, and delivering food to deep ocean ecosystems. The aim of GLOBESINK is to improve our ability to answer the big question: "How will storage of carbon by sinking ocean particles change in the future?". The answer could have wide-reaching impacts on climate, fisheries, and biodiversity. GLOBESINK will attack this big question using a new "big data" approach. The first step will be to dramatically increase the number of sinking particle flux measurements in the ocean, by approximately 10-fold. GLOBESINK will achieve this by applying its own innovative statistical methods to 10s of millions of particle measurements made by a new global fleet of over 200 drifting underwater robots. Using this new dataset, GLOBESINK will directly calculate the total global carbon stored and how this amount has varied in time and space over the past decade. This information will be combined with key measurements of particle size, rate of particle breakup, temperature, and oxygen concentration, all hypothesised to be important drivers of carbon storage, in order to assess whether and how these drivers should be included in global climate models.